Reduction of bis (cyclooctatetraene) uranium

File failed to load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js. ADVERTISEMENT · Log In Register · Cart · ACS · ACS Publicatio...
4 downloads 0 Views 227KB Size
J. Am. Chem. SOC.1981, 103, 5603-5604 approaches are complementary because they are applicable to molecules of different size. Deuterium N M R spectroscopy is most appropriate for small paramagnetics because the dominant linebroadening mechanism is quadrupolar relaxation9and is minimized by fast rotational correlation times. On the other hand, proton spectra are broadened by electron spin-nuclear spin dipolar relaxation, and the rapid reduction in line width with distance from the metal center (1/ f l ) l o favors this approach for large molecules. In the present example the deuterium quadrupole probe is preferred, particularly since the added mobility of the methyl group increases the deuterium T2. It is important to realize that dipolar coupling is inherent in all magnetically anisotropic molecules, and proper account of this effect is necessary when interpreting scalar spin-spin couplings. Since the order parameter is quadratic with field strength, observation and reporting of these effects will become more commonplace as superconducting magnet technology achieves higher fields. The value of such observations is clearly the structural information contained in the geometric parameters (ai,,Rij);our trivial example substantiating a methyl group geometry was chosen for simplicity. We expect field dependent studies of dipolar couplings will be applicable to structure determination in larger biological molecules with paramagnetic centers and can be generalized for any interacting I = 1/2 nuclei. For example, a typical paramagnetic with an anisotropy of 2000 X 10" cgs units mol-' will produce a dipolar coupling of 4.5 Hz between two protons separated by 2 A in a 600-MHz spectrometer. Acknowledgment. We are grateful for the skilled technical assistance of George Watunya and Bob Hibberd. This work was supported by a generous instrument grant from N I H RR00292. (9) Farrar, T. C.; Becker, E. D. "Pulse and Fourier Transform N M R ; Academic Press: New York, 1971; p 58. (IO) Swift, T. J. NMR Parumagn. Mol. 1973, 53-83.

Reduction of Bis(cyc1ooctatetraene)uranium F. Billiau, G. Folcher,* H. Marquet-Ellis, P. Rigny, and E. Saito Centre d'Etudes NuclZaires de Saclay DZpartement de Physico- Chimie 91 191 Gif-sur- Yvette, France Received May 1 , 1981

Uranium organometallics have been characterized by five different types of *-bonded ligands, viz., cyclopentadienyl, indenyl, cyclooctatetraene, allyl and arene. So far, of these five, only cyclopentadienyl and indenyl have been shown to occur in complexes of uranium(III).1-2 A possible route to U'" organometallics with other *-bonded ligands could be by the reduction of allylic derivatives. However, this could be expected to be difficult due to their high instability at room temperature. On the other hand, the synthesis of cyclooctatetraene derivatives (COT = C&) seems more feasible. The compound U1V(COT)2is thermally stable and soluble in quite a number of organic solvent^.^ Its sandwich structure, analogous to that of ferrocene, is found also in the case of lanthanum(II1) as in the anion Ln111(COT)2-.4 In the case of the reduction of U"(COT)2 many experiments have been undertaken but were hitherto unsuccessful, due certainly to an excess of reductant leading to an Uo c o m p o ~ n d . ~It is quite probable that (1) Kanellakopoulos, B.;Fischer, E. 0.;Dornberger, E.; Baumgartner, F. J. Organomet. Chem. 1970, 24, 540. (2) Manriquez, J. M.; Fagan, P.J.; Marks, T. J.; Vollmer, S. H.; Day, C.; Day, V. J. Am. Chem. SOC.1979, 101, 1057. (3) Streitweiser, A.; Muller-Westerhoff, U. J. Am. Chem. Sor. 1968, 90,

5603

the particular stability of U1V(COT)2and the reducing power of U1'' are responsible for the high fragility of the cyclooctatetraeneuranium(II1) compounds. We have obtained the reduction of U1V(COT)2to a U"' compound by controlling the amount of reducing agent used. The reaction was followed by observing the optical spectra, and the products were characterized by their 'H N M R signals. We have also obtained the same compound in solution by a second method using the reaction of U"'Cl,.THF with K2COT. The two methods mentioned above are as follows: (1) An aliquot volume of lithium naphthalenide in T H F was gradually added to the U1V(COT)2solution. The progress of the reduction was followed spectrophotometrically by observing the decrease of the four peaks of U'V(COT)2 (613,639,657, and 675 nm). The final clear brown solution had a spectrum which did not show any well-defined maxima in the visible range but presented a small band in the near infrared range (- 1700 nm) (Figure 1). It is worthwhile to note that the four bands of the initial U'V(COT)2 completely disappeared when we employed a CloH8-/U1" ratio of about 1. In such a case ESR measurements showed that no naphthalenide was left over after the reduction. This was also confirmed by the disappearance of the Cl,-J3f optical -800 nm (c 2.25 X lo3 L mol-'/cm-')] and the spectrum A[, appearance of the UV spectrum of naphthalene (310,285 nm). The 'H N M R spectrum of this solution presents a single broad (40 Hz) line at 37 ppm high field relative to CsHs at room temperature. It was possible to obtain crystals when a very pure solution of U(COT)2 was reduced by Li+CloH8- and kept overnight. The brown needles were extremely sensitive to air (pyrophoric) and are insoluble in toluene. Elementary analysis did not provide valid results owing to the high instability of the solid. (The C/Li ratio found was -23 and 1; the amount of product was insufficient to allow uranium C/H analysis.) Mass spectra present peaks which were assigned as follows: 446, U(COT),+; 151, Lb2THF; 78, Li-THF; 72, THF; 104, CgH8+. Peaks at 135 (CloH8Li),128 (CloHg), 1107, and 862 were also detected. The IH N M R signal of these crystals, very soluble in THF-d8, is identical with that obtained with the original solution (37 ppm; see above). This single N M R signal, of the same order of magnitude as that for U(COT)2, requires a sandwich structure in which the two COT rings are equivalent. The N M R and mass spectra strongly suggest the formula U(C8H8)2-Li,+.THF. (2) The same compound can also be obtained directly from U"'. To a solution of UC146in THF, an excess of N a H or LiH was added so as to obtain UC1,. THF6 identified by its visible spectrum (450, 475, 510, and 540 nm). To this solution was then added K2(COT), and the lH NMR signal of the resulting brown solution was identical with that obtained by method 1. That the uranium is trivalent has been well established as follows: (a) Figure 2 shows the ESR spectrum of a frozen solution cooled to 6 K. The similarity of this spectrum with those of other UI1I compounds7 which are very broad and asymmetric as well as the fact that Uvl and UIVdo not give ESR spectra and Uv does, but something quite different, leads us to assign our spectrum to U"'. This signal disappears at higher temperature or when the solution is oxidized. (b) The 'H NMR spectra show that the protons of the ligands COT undergo a paramagnetic shift of the order of 40 ppm at room temperature. Although the paramagnetic shifts induced by U"' are very close to those of U'", at least in the case of cyclopentadienyls,8 the effect of temperature on this shift shows a behavior quite different from that of UIVwhich follows the Curie-Weiss law (Figure 3). (c) Addition of this trivalent compound (obtained by using the first method) to a solution of TiIV(GOC3H7)4in THF leads to Ti"'

-

7161.

(4) Hodgson, K.0.; Mares, F.; Starks, D. F.; Streitweiser, A. J. Am. Chem. Soc. 1973.95, 8650. (5) Karraker, D. G.; Stone, J. A. J. Am. Chem. SOC.1974, 96, 6885.

0002-7863/8 1/1503-5603$01.25/0

(6) Moody, S. J. Inorg. Nucl. Chem. 1979, 41, 533. (7) Soulie, E.; Folcher, G.; Marquet-Ellis, H., to be published. (8) Folcher, G.; Marquet-Ellis, H. J. Orgunomet. Chem. 1977, 257, 131.

0 1981 American Chemical Society

5604

J. Am. Chem. SOC.1981, 103, 5604-5606 A

-

I

I

I

I

I

300

400

500

600

700

-

I

BOO nm

Figure 1. Spectrophotometrical study: (a) Absorption spectrum of the initial solution of U*V(COT)2in THF (agent (Np-/U'" naphthalene.

0.4). (c) Absorption spectrum of the U(II1) compound (Np-/UIV

Figure 2. ESR spectrum of a frozen solution cooled to 6 K.

M). (b) After addition of reducing 1) with its IR component (a'). (d) Absorption spectrum of

In the same way, instead of a TiIV compound, we can use UrV(CsHS)3Clwhich is reduced to U111(CSH5)3 while the UInCOT compound is oxidized to UrV(COT)2. The U(CSHs)3is identified by its chemical shift of 20 ppm.* This shows that the U"' has more reducing power when ligated to COT than to C HS. This study has shown that it is possible to reduce U1'(COT)2 to a cyclocctatetraenyluranium(II1) compound in a T H F solution by using a stoichiometric amount of a strong reducing agent. The UII1compound, whose structure has still to be determined with certitude, is very soluble in T H F and has reducing properties toward Ti(OR)4 and U(CSHS)$l.

Ptilocaulin and Isoptilocaulin, Antimicrobial and Cytotoxic Cyclic Guanidines from the Caribbean Sponge Ptilocaulis aff. P. spiculifer (Lamarck, 1814)'

10.30

Gary C. Harbour, Adrienne A. Tymiak, and Kenneth L. Rinehart, Jr.* Department of Chemistry, Roger Adams Laboratory University of Illinois at Urbana- Champaign Urbana, Illinois 61801 Paul D. Shaw Department of Plant Pathology University of Illinois at Urbana-Champaign Urbana, Illinois 61801 T 0

300' K (A in ppm) vs. temperature (K). (0)values for 200°

100"

Figure 3. l/A, u(IV);* (x)our values for u(II1).

with its characteristic ESR signal possessing hfs due to 47Ti ( I = 5/2) and 4?i ( I = 7/2) and a g value of 1.971? In this reaction the uranium compound is oxidized to U1V(COT)2as shown by the four absorption peaks. (9) Saito, E.; Billiau, F., to be published.

Robert G. Hughes, Jr. Department of Cell and Tumor Biology Roswell Park Memorial Institute Buffalo, New York 14263 Stephen A. Mizsak, John H. Coats, Gary E. Zurenko, Li H. Li, and Sandra L. Kuentzel The Upjohn Company Kalamazoo, Michigan 49001 Received April 13, 1981

In our recent report2 on the most bioactive extracts obtained during the Alpha Helix Caribbean expedition 1978, we noted that

0002-7863/81/1503-5604$01,25/00 1981 American Chemical Society